Professional geostatistical system for archaeology - A complete QGIS plugin for advanced spatial analysis, multivariate kriging, machine learning, and optimal sampling design for archaeological research.
- Variograms with weighted model fitting (spherical, exponential, Gaussian, Matérn) and directional anisotropy check
- Ordinary Kriging with uncertainty map and leave-one-out cross-validation (RMSE, MAE, ME, R², MSDR)
- Co-Kriging (regression + residual kriging), spatio-temporal kriging and batch kriging in the Processing Toolbox
- Compositional analysis (CLR/ILR/ALR transforms)
- Spatial clustering (K-Means, DBSCAN) and anomaly detection (Isolation Forest) in the panel
- Random Forest supervised classification (Processing, with a label field)
- Native QGIS Processing Provider
- GeoTIFF rasters (prediction + uncertainty) added to the project
- Interactive charts with Plotly (static preview when QtWebEngine is missing)
- Italian/English interface, built-in tutorial and example data
- Ceramic analysis and artifact distribution
- GPR + Magnetometry fusion
- Compositional soil analysis (CLR/ILR transforms)
- Optimal sampling design
# Python dependencies
pip install numpy scipy pandas scikit-learn plotly
QGIS already ships numpy, scipy and (usually) scikit-learn, plotly and matplotlib; the plugin offers to install anything missing on first start.
- Download the latest release: Download GeoArchaeo.zip
- In QGIS:
Plugins > Manage and Install Plugins > Install from ZIP - Select the downloaded ZIP file
- Restart QGIS
# Linux/macOS
cd ~/.local/share/QGIS/QGIS3/profiles/default/python/plugins/
git clone https://github.com/enzococca/GeoArchaeo.git
# Windows
cd %APPDATA%\QGIS\QGIS3\profiles\default\python\plugins\
git clone https://github.com/enzococca/GeoArchaeo.gitThe plugin is available in the official QGIS Plugin Repository.
- Open QGIS and click the GeoArchaeo icon in the toolbar (or load the example data from the panel).
- At the top of the panel choose the point layer and the numeric field once: every tab uses them.
- Data tab: statistics and histogram.
- Variogram tab: "Auto" distance, spherical model, Calculate variogram.
- Kriging tab: check the grid size shown, Run kriging → prediction and uncertainty rasters are added to the map, with cross-validation metrics.
- Report tab: save an HTML report with all the charts.
import processing
# Variogram
result = processing.run("geoarchaeo:variogram", {
'INPUT': 'path/to/points.shp',
'FIELD': 'ceramic_count',
'MAX_DISTANCE': 50,
'MODEL': 0, # Spherical
'OUTPUT': 'variogram.json'
})
# Ordinary Kriging
kriging = processing.run("geoarchaeo:ordinarykriging", {
'INPUT': 'path/to/points.shp',
'FIELD': 'ceramic_count',
'PIXEL_SIZE': 2.0,
'OUTPUT': 'kriging_result.tif'
})id,x,y,type,quantity,period
1,345678.5,4567890.2,ceramic,25,roman
2,345679.1,4567891.7,metal,5,romanRequirements:
- Coordinates in a projected CRS (e.g. UTM)
- At least one numeric field for interpolation
- Minimum 30 points for robust analysis
graph LR
A[Import Data] --> B[Quality Check]
B --> C[Variogram]
C --> D[Kriging]
D --> E[Validation]
E --> F[Export Maps]
Identify functional areas (kitchens, storage rooms) based on fragment distribution.
Combine GPR + Magnetometry to identify buried structures.
CLR/ILR transforms for granulometric and chemical soil data.
Calculate optimal positions for new test trenches by minimizing uncertainty.
The repository includes sample datasets for testing:
test_layers/
├── villa_ceramica.gpkg # 500 points - ceramic distribution (EPSG:32633)
├── geofisica_survey.gpkg # 4800 points - GPR + magnetometry
├── necropoli.gpkg # 120 tombs with attributes
├── soil_samples.gpkg # 80 points - compositional analysis
├── area_studio.geojson # study-area polygon for sampling design
└── *.csv # the same data and extra surveys as CSV (x, y columns)
Use Load example data in the panel (or the toolbar button) to add them to the project.
geoarchaeo/
├── __init__.py # Entry point
├── geoarchaeo_plugin.py # Main plugin class
├── processing_provider.py # Processing algorithms
├── compat.py # Qt5/Qt6 compatibility
├── i18n.py # Internationalization (IT/EN)
├── core/
│ └── geostat_engine.py # Geostatistical engine (QGIS-independent, vectorised)
├── gui/
│ ├── main_dock.py # Dock widget interface
│ └── tutorial_html.py # Built-in tutorial (IT/EN)
├── tests/ # pytest suite (engine + headless QGIS)
├── icons/ # Icons and resources
├── doc/ # Documentation
├── test_layers/ # Sample datasets
└── metadata.txt # QGIS metadata
# engine only (any Python with numpy/scipy/scikit-learn)
python -m pytest tests/test_engine.py
# full suite with the QGIS Python (macOS example)
/Applications/QGIS.app/Contents/MacOS/bin/python3 -m pytest tests/Contributions are welcome! To contribute:
- Fork the repository
- Create a branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
If you use GeoArchaeo in your research, please cite:
@software{geoarchaeo2024,
author = {Cocca, Enzo},
title = {GeoArchaeo: Advanced Geostatistical Analysis for Archaeological Research},
year = {2024},
publisher = {GitHub},
url = {https://github.com/enzococca/GeoArchaeo}
}- Bug reports: Open an issue
- Email: enzo.ccc@gmail.com
This project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details.
- QGIS Development Team for the excellent platform
- Archaeological community for feedback and testing
- Complete geostatistical core
- Processing integration
- Basic Machine Learning
- Full documentation
- Plugin in official QGIS repository
- Bilingual GUI (IT/EN)
- More languages (ES/FR)
- Video tutorials
- Online workshops
- Repository: github.com/enzococca/GeoArchaeo
- Issues: Bug Reports
- QGIS: qgis.org
Developed for the archaeological community by Enzo Cocca
